-
1
Project title: Understanding and developing methods for
managing spotted wing drosophila (SWD) in the UK:
Vital research to maintain the viability of the UK fruit
industry
Project number: SF145
Project leader: Michelle Fountain/ Jerry Cross, East Malling
Research, New Road, East Malling, Kent ME19 6BJ
Report: Annual report, March 2016, Year 3
Previous report: Years 1 and 2
Key staff: Prof Jerry Cross, Dr Michelle Fountain, Dr David
Buss, Bethan Shaw, Janie Morton, Katie Hopson,
Dieter Baets, Maddie Cannon, Adrian Harris, Alvaro
Delgado, Emilio Fernandez Rico, Dr Ralph Noble,
Andreja Dobrovin-Pennington, Phil Brain (NIAB
EMR)
Prof David Hall, Dudley Farman (NRI)
Dr Alison Dolan (JHI)
Key collaborators Berry Gardens, Horticultural Development
Company
(HDC), James Hutton Institute (JHI), Natural
Resources Institute (NRI), Worshipful Company of
Fruiterers
Location of project: NIAB EMR
Industry Representative: Marion Regan, Hugh Lowe Farms
Date project commenced: 01 April 2013
Date project completed
(or expected completion
date):
31 March 2017
-
2
DISCLAIMER
While the Agriculture and Horticulture Development Board seeks
to ensure that the information
contained within this document is accurate at the time of
printing, no warranty is given in respect
thereof and, to the maximum extent permitted by law the
Agriculture and Horticulture Development
Board accepts no liability for loss, damage or injury howsoever
caused (including that caused by
negligence) or suffered directly or indirectly in relation to
information and opinions contained in or
omitted from this document.
© Agriculture and Horticulture Development Board [2016]. No part
of this publication may be
reproduced in any material form (including by photocopy or
storage in any medium by electronic
mean) or any copy or adaptation stored, published or distributed
(by physical, electronic or other
means) without prior permission in writing of the Agriculture
and Horticulture Development Board,
other than by reproduction in an unmodified form for the sole
purpose of use as an information
resource when the Agriculture and Horticulture Development Board
or AHDB Horticulture is clearly
acknowledged as the source, or in accordance with the provisions
of the Copyright, Designs and
Patents Act 1988. All rights reserved.
All other trademarks, logos and brand names contained in this
publication are the trademarks of
their respective holders. No rights are granted without the
prior written permission of the relevant
owners.
[The results and conclusions in this report are based on an
investigation conducted over
a one-year period. The conditions under which the experiments
were carried out and the
results have been reported in detail and with accuracy. However,
because of the
biological nature of the work it must be borne in mind that
different circumstances and
conditions could produce different results. Therefore, care must
be taken with
interpretation of the results, especially if they are used as
the basis for commercial
product recommendations.]
-
3
AUTHENTICATION
We declare that this work was done under our supervision
according to the procedures
described herein and that the report represents a true and
accurate record of the results
obtained.
Michelle Fountain
Research Leader in Entomology
NIAB EMR, New Road, East Malling, Kent ME19 6BJ
Signature
............................................................ Date
.........11 March 2016......
Jerry Cross
PPESCM Programme Leader
NIAB EMR, New Road, East Malling, Kent ME19 6BJ
Signature
............................................................ Date
........ 15 Mar 16............
Report authorised by:
[Name]
[Position]
[Organisation]
Signature
............................................................ Date
............................................
[Name]
[Position]
[Organisation]
Signature
............................................................ Date
............................................
-
4
GROWER SUMMARY
Headline
A greater understanding of the biology and control of SWD in the
UK has been achieved with
findings directly relevant to UK soft and stone fruit
growers.
Background and expected deliverables
Spotted wing drosophila (Drosophila suzukii, SWD) is a new
invasive pest to the UK, but has
caused considerable losses in fruit crops in Europe and the USA.
The overall aim of the project is
to monitor the spread of D. suzukii within the UK and to develop
measures for its control. To this
end five objectives have been set for the project:
1. To determine the distribution and seasonal population
dynamics of all life stages of D. suzukii
in different cropping situations and especially polytunnel crops
on fruit farms in the UK.
2. To develop economically and environmentally sustainable
treatment and disposal strategies
for soft and stone fruit waste to eliminate it as a source of D.
suzukii infestation and attraction
on fruit farms.
3. To develop and evaluate sampling and extraction methods for
quantifying D. suzukii
infestations in different soft and stone fruits.
4. To develop a synthetic lure and attract and kill technology
for D. suzukii for incorporation into
IPM programmes.
5. To obtain evidence for the effectiveness of different plant
protection products including
biopesticides to aid developing an insecticide resistance
management strategy for SWD.
-
5
Summary of the project and main conclusions
Objective 1: To determine the distribution and seasonal
population dynamics of all life stages of Drosophila suzukii in
different cropping situations and especially polytunnel crops on
fruit farms in the UK and investigate its wide wild hosts and
overwintering.
National monitoring
In 2015, the national monitoring of adult D. suzukii numbers was
continued at a network of 15 sites
across the UK using modified Biobest traps with Cha-Landolt
bait: 5 in Kent, including NIAB EMR,
1 in Surrey, 3 in the West Midlands, 2 in East England and 4 in
Scotland, including the James
Hutton Institute.
Numbers of D. suzukii caught were considerably higher in 2015
than in previous years. The largest
catches were in the south east of England, but D. suzukii were
found at all sites in 2014. In a
similar pattern to previous years the numbers caught in crops
peaked in August before declining
and then rose again in late October, with adults migrating to
woodlands and hedgerows where
numbers remained high throughout November and December, reaching
a peak of over 20,000 per
trap per week at one site. It is estimated that numbers in the
winter of 2015 are at least 3 times
higher than at the same time in 2014, although the winter has
been quite mild, probably resulting
in a higher activity and trapping efficacy.
Habitat survey
The distribution of D. suzukii on two farms, including NIAB EMR
was studied throughout the year.
Over 50 traps were deployed on each farm in a range of crops and
in neighbouring wild areas and
woodlands. D. suzukii were recorded in the traps every week in
2015, in contrast to 2014 when
for 5 weeks in the spring none were recorded.
The trap catch throughout 2015 showed a similar pattern to
previous years. Activity increased in
early spring from 14 April to 8 June, with more pronounced
activity towards the end of the summer
from 27 July until the winter. Numbers peaked in late autumn and
winter with considerably higher
catches than in 2014. Higher numbers of D. suzukii were caught
in cherry orchards than in other
crops surveyed. The number of D. suzukii decreased in the cherry
orchards once all the leaves
had fallen.
All reproductive stages were seen during the fruit growing
season on both farms in 2015. D.
suzukii from Farm 2 appeared to have a longer period of
fertility than Farm 1 (April through to
-
6
November), a trend that was found in the previous year. This may
be related to the later removal
of polytunnels and/or the removal of raspberry waste at Farm
1.
Population modelling
A computer model was constructed using fertility studies,
climatic data and biological parameters
derived from this project and from published literature, to
predict D. suzukii population changes.
This model is being validated against population data collected
during the course of this project
and gives a good estimate of the time of first egg laying and
the start of population growth. Future
research should continue to optimise the programme, especially
for population prediction later in
the year.
Monitoring SWD larval infestations in early, mid and late season
cherry varieties
In the first trial, seven commercial varieties of cherries were
assessed covering early to late
cropping. Brix (sugar content), hardness and D. suzukii
emergence were monitored for each crop.
In general there was a positive relationship between Brix and
the softening of the cherry fruits,
especially Sweetheart, Korvic and Merchant.
D. suzukii larvae were not found in the early variety Simone
(picked first week of July) and although
they emerged from the other early variety, Merchant, this was
only after the optimal time for harvest
had passed and the fruits were very soft. If no plant protection
products were applied eggs could
be laid in Sweetheart and Penny very early.
In a second survey of D. suzukii occurrence in four commercial
orchards in Southern England, no
D. suzukii emerged from fruits harvested between 14 May and 8
June, however small numbers
were recorded after this time.
UK host plants for D. suzukii and overwintering sites.
Black Bryony and Yew berries were shown to be potential
oviposition sites towards the end of the
year. However, Cotoneaster, Snowberry, Guelder Rose, Dogwood,
Hawthorn, Red Bryony and
Rose do not appear to support D. suzukii development. All of
these fruits are rather dry and fibrous.
Raspberry cuttings and leaf litter did not appear to be
significant sheltering sites for D. suzukii in
the late autumn and winter in this limited study.
-
7
Objective 2. To develop economically and environmentally
sustainable treatment
and disposal strategies for soft and stone fruit waste to
eliminate it as a source of
D. suzukii infestation and attraction on fruit farms.
Treatment in anaerobically sealed Dolav bins was shown to kill
all D. suzukii in soft fruit waste, as
long as the ambient temperature was over 18oC. However, if the
ambient temperature was lower,
such as might be found at the end of the season, then three days
would be necessary to kill eggs
and larvae.
Stone fruit was shown to take a longer period of storage to
remove all D. suzukii, so that four days
of storage are required, or five days if temperatures are below
16oC ambient temperature.
Oxygen depletion was very rapid for each fruit type. It was
non-detectable after 6 hours and it is
possible that this is the crucial factor in killing eggs and
larvae. There was a rapid increase of CO2
levels in soft fruit waste, but this was much slower in stone
fruit waste, possibly because of the
firmer nature of the fruit and the presence of air pockets.
Mixing treated waste with at least 90% (w/w) organic matter such
as manure or slurry was shown
to prevent re-inoculation, as was rotavation to a depth of 20
cm. The rate of application of treated
waste to land should not exceed 125 tonnes/ha to prevent
exceeding EU directives on nitrate
addition.
Disposal of fruit waste via digestion plants was not considered
to be financially attractive due to
the high moisture content and low calorific value of fruit
wastes, along with transport and gate fee
costs.
Objective 3. To develop and evaluate sampling and extraction
methods for
quantifying D. suzukii infestations in different soft and stone
fruits.
Low cost methods were trialled to detect D. suzukii larvae (both
early and late stage) in samples
of blueberries, cherries, raspberries and strawberries. The
methods included immersion of
crushed fruit in strong sugar or salt solutions, in a weak
detergent solution or freezing whole fruit
overnight. These methods were compared to emergence testing
(keeping fruits in boxes at room
temperature for 3 weeks and counting adult emergence) and
dissecting the fruits open to count
the numbers of larvae directly by hand. Sugar and salt immersion
were the most successful in
detecting D. suzukii larvae, whether late or early stage, with
sugar solution slightly more effective.
No method gave 100% recovery of the larvae.
-
8
Flotation with a strong sugar solution was the most practical
way to determine the infestation levels
of fruits and a standard protocol has been prepared in
conjunction with AHDB Horticulture for
growers. Training videos and training posters on how to
undertake a flotation test using sugar
solution have been published by AHDB Horticulture and are
available on the dedicated SWD site
of the AHDB Horticulture website.
Evaluating intra species competition for egg laying sites
D. suzukii appear to prefer to oviposit into media that has not
been exposed to D. melanogaster.
The latter do not seem affected by previous exposure to D.
suzukii egg laying. This may be related
to the niche that D. suzukii occupies; they do not need to
compete with other Drosophila species
for egg laying sites, whereas D. melanogaster need to compete
for egg laying sites with other UK
species. The absence of competition may also explain the
relatively long development time of D.
suzukii.
Objective 4. To develop a synthetic lure and attract and kill
technology for D. suzukii
for incorporation into IPM programmes
Testing commercially available and experimental baits
A comparison of various commercially available D. suzukii lures
and traps was made, comparing
D. suzukii catch, bycatch and ease of use. A detailed breakdown
of the results is provided in the
Science Section of this report. It was also found that topping
up precision monitoring traps with
DrosAttract rather than replacing the whole unit was as
effective and cheaper, whilst producing
less bycatch and waste. Chemical analysis of the successful
commercial baits suggested that
ethanol might not be as necessary as previously believed. In
contrast, 3-Methylbutanol was
released from all baits tested and might be a useful component
of future lures.
Develop target device and identify suitable insecticide(s) for
attract and kill formulation
An ‘attract and kill’ device is being developed in conjunction
with the NRI. Miniature, dry versions
of the Cha Landolt lures have been shown to be effective in
attracting D. suzukii and deltamethrin
(Decis) appears to be a good candidate choice for an
insecticide. This work will be continued in
2016.
-
9
The evaluation of components of Cha-Landolt baits for the
efficiency of trapping
Delivery systems for the components of the Cha Landolt trap
system, ethanol, acetic acid, acetoin
and methionol were assessed. Sachets for acetoin and methionol
developed by NRI were found
to be as effective as the vials previously used and much easier
to use. However, ethanol and
acetic acid were more effective in the drowning solution than in
sachets or vials. This is being
investigated further.
Objective 5. To obtain evidence for the effectiveness of
different plant protection
products including biopesticides and for developing an
insecticide resistance
management strategy for D. suzukii.
Evaluate the efficacy of approved and emerging products against
adults and other life stages in
crops
In a replicated field experiment, with a population of D.
suzukii, crop protection products were
assessed for effectiveness of D. suzukii control on cherries.
The efficacy of D. suzukii control
varied with the plant protection product applied and time post
spraying. Spinosad, lambda
cyhalothrin, cyantraniliprole and a coded product gave good
control over the duration of the study,
whilst deltamethrin, acetamiprid and another coded product gave
good initial protection, but by
day 14 were beginning to lose effectiveness. In contrast, one
application of lime and a pyrethrins
mixture gave relatively poor control.
Evaluate the use of sugar as an adjuvant for enhancement of
insecticide treatments in the control
of D. suzukii
Sugar was investigated as a way of enhancing the effectiveness
of plant protection products
against D. suzukii. A literature review found sugar to be
effective in a number of cases. In our
trials, sugar significantly enhanced adult mortality from
chlorantraniprole. However, no significant
effect of sugar on adult mortality or subsequent emergence was
found with spinosad, lambda
cyhalothrin or deltamethrin.
-
10
Additional research
Efficacy of Jet 5 or Lime for the control of egg laying of
Drosophila suzukii
Jet 5 and lime were assessed for usefulness in D. suzukii
control. Blueberries were dipped in the
treatments and adult flies were added. Direct mortality, egg
laying and subsequent emergence of
adults from the fruit were measured. Jet 5 caused mortality to
flies added to the fruit, but this
may be due to vapour action within the experimental arena. Lime
reduced the number of flies
emerging from treated fruit indicating a repellent effect on egg
laying
Population dynamics of Drosophila suzukii in relation to other
Drosophila species in the UK
2014-17
Six Drosophila species and the species group Drosophila obscura
were regularly found in the
national monitoring traps. These species may interact with D.
suzukii via competition, the
transmission of pathogens or spread of parasitoids. The
commonest bycatch was the D. obscura
group. These were found in all sites and all time periods. In
terms of monitoring, bycatch of other
Drosophila can slow assessment considerably and several species
can look like D. suzukii to an
inexperienced observer.
Financial benefits
D. suzukii poses a clear threat to the fruit industry and has
had a commercial impact on UK grown
fruit since 2014. Growers have reported significant financial
losses in cherry and some soft fruit
crops.
-
11
Action points for growers
Monitor adults in susceptible crops and wild areas around crops
from February onwards to
predict the onset of egg laying by D. suzukii. Use a recommended
trap and bait.
Consider winter trapping and deploy perimeter trapping around
vulnerable crops before fruit
begins to ripen, potentially delaying movement of D. suzukii
into the crop.
Monitor for the presence of larvae in the crop. The flotation
technique using sugar solution is
recommended for rapid detection of larvae, but growers should
consider emergence testing
(boxes of fruit at room temperature) for early season
detection.
Crop hygiene is one of the best methods of population control.
All waste fruit should be
removed from the crop during and after harvest and be treated in
sealed vessels and then
disposed of responsibly by incorporating into the surface of
field soils or mixed with organic
matter such as manure or slurry.
Consult BASIS trained advisers for the latest approvals for
effective plant protection products
and use the comprehensive information on SWD in the dedicated
SWD site of the AHDB
Horticulture website http://horticulture.ahdb.org.uk/.
http://horticulture.ahdb.org.uk/
-
12
SCIENCE SECTION
Objective 1. To determine the distribution and seasonal
population dynamics of all
life stages of SWD in different cropping situations and
especially polytunnel crops
on fruit farms in the UK and investigate its wild hosts and
overwintering.
Task 1.1. Determine the population dynamics of adult SWD in
vulnerable polytunnel and
outdoor grown fruit crops at 13 sites in the different fruit
growing regions of England and
Scotland throughout the year for four successive years.
Materials and methods
Sites
Fourteen fruit farms selected in year one (2013) were used, with
an additional farm that was added
in the West Midlands from 2014 (Table 1.1.1). The distribution
of the farms were as follows; five
in Kent (including East Malling Research), one in Surrey, three
in the West Midlands
(Herefordshire and Staffordshire), two in Eastern England
(Northamptonshire and Norfolk) and
four in Scotland (including the James Hutton Institute). Farms
were chosen to give good
geographical coverage and to ensure that a full range of
vulnerable soft and stone fruit crops were
assessed; blackberry, blueberry, cherry, raspberry, redcurrant
and strawberry crops were included
in the trial sites. At least one woodland area was also assessed
at each farm.
-
13
Table 1.1.1. Summary of fruit farms involves in the national
monitoring survey (2015)
Region and crops
South East England (46 traps)
Farm 1 Raspberry, strawberry
Farm 2 Raspberry, strawberry
Farm 3 Cherry
Farm 4 Raspberry
Farm 5 Cherry, strawberry, plum, grape
Farm 6 Blueberry, redcurrant, strawberry
Eastern England (20 traps)
Farm 7 Blueberry, raspberry
Farm 8 Raspberry, strawberry
West Midlands (27 traps)
Farm 9 Blackberry, blueberry, raspberry, redcurrant,
strawberry
Farm 10 Blueberry, cherry, raspberry, strawberry
Farm
10a
Cherry, raspberry
Scotland (40 traps)
Farm 11 Blackcurrant, blueberry, raspberry, strawberry
Farm 12 Blueberry, cherry, strawberry
Farm 13 Blackberry, blueberry, raspberry, strawberry
Farm 14 Blackberry, blueberry, raspberry, strawberry
-
14
D. suzukii traps
Monitoring traps were deployed in pairs, one in the centre and
one at the edge of each crop. Pairs
of traps were also deployed in wooded areas on each farm. For
continuity, within the National
Monitoring scheme we continued to use the modified Biobest trap
design and Cha-Landolt bait
used at the end of 2013. Trials at EMR on comparing and
improving lure and trapping technology
for D. suzukii are described in detail under Objective 4.
Droso-traps (Biobest, Westerlo, Belgium) were modified with 20
extra 4 mm holes drilled into the
top portion of the body of the trap to maximise catches of D.
suzukii. Flies were caught in a
drowning solution, which also included ethanol (7.2%) and acetic
acid (1.6%) as attractants, and
boric acid to inhibit microbial growth. Methionol and acetoin
(diluted 1:1 in water) were released
from two polypropylene vials (4 ml) with a hole (3 mm dia.) in
the lid, attached near the entry holes
within the trap. The traps were deployed at the height of the
main crop. In strawberry fields, traps
were hung off the ground to prevent slugs entering the traps,
but low enough so that they passed
under the sprayer.
Trapping began in May 2013 and is being continued with weekly
counts during the cropping
season and biweekly counts during the winter.
Results
The results for England in 2015 are summarised in Fig. 1.1.1, in
comparison to 2013 and 2014.
The results for mid-February to mid-August are shown in more
detail in Fig. 1.1.2.
Overall, numbers in January 2015 were far higher than the
previous year, before falling to low
levels in early February. Catches of D. suzukii adults continued
at low levels, although higher than
2014, until early August, at which point catches increased
rapidly. This increase corresponded
closely to the pattern for 2014, including a temporary drop in
September, but from September
onwards numbers in 2015 were considerably higher than the
previous year.
There was considerable variation between sites within the UK
regions, but some trends are evident
(Fig. 1.1.3). As in 2014, the largest catches were in the south
east of England. Here adult D. suzukii
were caught in woodland areas in January and before falling to
very low numbers in all areas and
habitats in early-February. In March, April, May and June only
low numbers of D. suzukii appeared
in the traps, though there was evidence of a slight peak in
woodland traps in March. The numbers
caught in crops rose to a peak in August, before falling and
then rising again in September. At the
time of writing (early November) numbers are still
increasing.
-
15
Catches in traps in the East of England and the West Midlands
remain, on average, low compared
to the South East. After January, almost no adult D. suzukii
were caught until August. From then
on, trap catches in both regions followed a similar pattern to
the South East until October, though
at much lower numbers and without a discernible peak in August.
From October onwards numbers
declined.
No D. suzukii had been caught in the National Monitoring scheme
in Scotland in 2013, and the
first were detected at the end of July 2014. Numbers in 2015
remained very low compared to the
rest of the country, although catches were starting to increase
in October.
-
16
Figure 1.1.1. Comparison of average adult D. suzukii catch per
trap in 2013, 2014, 2015 and 2016.
Figure 1.1.2. Comparison of average adult D. suzukii catch per
trap from mid-February to mid-
August 2014 and 2015
-
17
Figure 1.1.3. Mean numbers of D. suzukii adults per trap in the
main habitat types and regions
during 2015
Discussion
Catches of D. suzukii were relatively high in January, at least
in the South East. This may be a
consequence of the warm November and December the previous year
and populations rising over
time since introduction and becoming more established (Fig.
1.1.4). It could also be a
consequence of higher woodland cover and hence a greater ability
to overwinter in the South East
(Fig. 1.1.5).
As in 2014, catches of adult D. suzukii in the traps were very
low in the UK until late July. It was
anticipated, given the higher overwintering population and
geographical spread in 2015, that
numbers in the spring and summer might be very high. However,
because traps are in competition
with fruits during the summer the traps are probably not
representative of ‘real’ in-field populations.
There was a sharp increase in the numbers caught in woodland in
the late autumn of 2015. This
pattern was also found in 2013 and 2014, as well as other
temperate regions at the northern limits
of the D. suzukii range, and presumably reflects a migration to
more sheltered areas. The numbers
0
1000
2000
3000
4000
5000
6000
7000
8000
Me
an N
o. S
WD
per
tra
pSouth East Crops
South East Woodland
East Crops
East Woodland
West Midlands Crops
West Midlands Woodland
Scotland Total
-
18
trapped were at least 3 times higher than the previous year. In
one trap, in one week, in a
woodland, at one farm there were 15,000 D. suzukii in December
2015.
Figure 1.1.4. Comparison of the weather in the South East and
Central South region of the UK in
2013, 2014 and 2015 with the 1980-2010 average (Met Office,
UK)
-
19
Figure 1.1.5. Percentage woodland cover (1998) from;
http://www.forestry.gov.uk/pdf/nigreatbritain.pdf/$FILE/nigreatbritain.pdf
Summary
Numbers of adult D. suzukii caught in traps were higher in 2015
than previous years
All trap sites caught D. suzukii
Largest numbers were caught in late autumn and winter,
especially in woodland areas.
http://www.forestry.gov.uk/pdf/nigreatbritain.pdf/$FILE/nigreatbritain.pdf
-
20
Task 1.2. Determine the phenology, population dynamics and
spatial distributions
of SWD on two fruit farms in SE England, including one
polytunnel cherry, raspberry
and strawberry crop, throughout the year for four successive
years (EMR; 1-4).
Materials and Methods
Adult trapping
Commercial polytunnel crops (cherry, raspberry and strawberry)
were assessed at two farms in
Kent. In addition, a variety of surrounding habitats were
included for analysis including woodland,
hedgerow and compost heaps. Traps were distributed to cover the
edges and centre of the focus
crops.
Twenty seven pairs of traps were originally deployed on Farms 1
and 2 in 2013. The strawberry
crop was removed in October 2014 resulting in the traps being
moved to a strawberry crop on
Farm 1. In 2015, the number of traps were reduced to twenty
three as key areas of interest had
been identified, reducing the need for duplicate locations.
Traps were removed from wasteland
and pear and apple orchards leaving the sites described in Table
1.2.1. The traps used were
based on the Cha-Landolt system (as used in the National
Monitoring scheme) consisting of two
polypropylene vials, one acetoin and one methionol, and a
synthetic bait (ethanol, acetic acid,
boric acid and detergent) which was also used as a drowning
solution to capture insects entering
the trap (Fig. 1.2.1). Each location was allocated two traps
which were spaced 10 metres apart.
Figure 1.2.1. Modified Drosotraps with synthetic bait
Trap assessments were completed weekly from February until
December in 2015. Trap contents
were taken back to the laboratory to be assessed, where male and
female D. suzukii were counted
plus any other Drosophila trapped in the solution. From each
trap two male and two female D.
suzukii were kept and stored in 70% alcohol for further
analysis.
-
21
Table 1.2.1. Numbered pairs of traps at Farm 1 and 2 and
associated
habitat.
Farm 1 Farm 2
Trap pair no. Habitat Trap pair no. Habitat
1 Cherry orchard 113 Raspberry
2 Cherry orchard 114 Raspberry
4 Woodland 115 Raspberry
5 Cherry orchard 116 Woodland
9 Cherry orchard 118 Woodland
10 Cherry orchard 119 Raspberry
13 Cherry orchard 127 Raspberry
14 Cherry orchard 128 Raspberry
16 Hedgerow
19 Strawberry
24 Compost heap
25 Strawberry
26 Strawberry
27 Strawberry
28 Strawberry
Habitat assessments
Trap catches were assessed in groups defined by habitat type.
Records of plant species diversity
and abundance were taken from the areas surrounding the paired
traps. Abundance was
calculated using the Total Estimate Scale (Table 1.2.2). The
growth stage of each crop was
recorded using definitions published by the European and
Mediterranean Plant Protection
Organisation (EPPO). Wild fruits were assessed by observing
percentage of ripeness, divided
into under ripe, ripe and over ripe.
Results from the adult trapping were correlated with their
habitat type. Data was correlated weekly.
Monthly habitat assessments recorded the presence of wild and
cropping fruiting plant abundance
and growth stage.
-
22
Table 1.2.2. The Total Estimate Scale
Fruit samples
Samples of fruit were taken weekly from both the edges and
centre of each focus crop, from the
green fruit stage until the end of fruiting. Within each soft
fruit crop 50 fruit were sampled between
the pair of traps. Fruits picked were low in the canopy
(raspberry) and overripe or damaged looking
(where possible). For the cherries, a sample was taken from each
of the six main cherry varieties
(confidential), adjacent to where the traps were placed. Fruit
samples were put in ventilated
Perspex boxes at 26°C for three weeks with 16:8 h light: dark
regime. The boxes were checked
twice weekly for three weeks to test for the presence of D.
suzukii adults (emergence testing).
An additional ten cherries were sampled from each crop and
measurements of colour (comparison
with Ctfil colour chart), hardness/softness (measurements with
penetrometer/durometer) and
sugar content (refractometer measurements of °Brix) made weekly
from June until the end of
harvest in July.
Fruit samples were checked weekly for the presence of D. suzukii
adults (emergence testing).
Flies that did emerge were removed and counted.
Reproductive stage of females
Trapped female D. suzukii were stored in 70% alcohol and
refrigerated. Females were dissected
under a dissecting microscope with light source. Measurements of
the body length, and wing
length and depth were taken using a graticule within the eye
piece. General observations of the
colouration and banding depth were recorded using a similar
system to that of G. Petavy et al.
(2002) which looks at the number of darkened tergites. The
abdomens of the females were
removed by grasping the thorax in one set of forceps and the
ovipositor in another and pulling
apart.
Assessments of females were made weekly for those dates in which
females were collected from
the two farms in the habitat assessments. Five females per
sample were randomly taken and
floated in 70% alcohol within a petri dish. The state of
reproductive maturity is derived from a
visual assessment of ovary and egg development using stage
definitions published by Beverly S.
Gerdeman, Washington State University.
-
23
Results
Trapping
Trapping continued through the winter of 2014/15. D. suzukii
were recorded in the traps every
week of 2015. The previous year in comparison had five weeks
where no D. suzukii were recorded
in the traps. Farm 1 had three weeks (07/02/14, 07/03/14
&13/06/14) and Farm 2 had two weeks
(14/02/14 & 06/06/14) where the traps were completely devoid
of D. suzukii (Fig 1.2.2). However,
in 2014 and 2015 trap catches show a distinct pattern whereby
the number of D. suzukii are
considerably less at the beginning of the year, suggesting that
the population of D. suzukii active
at this time is most likely in their overwintering forms laying
the first generation eggs for when
habitat conditions are more favourable (Fig 1.2.2).
Figure 1.2.2. Adult D. suzukii trap catches at farms 1 and 2 in
2013, 2014 and 2015.
The rapid increase in activity between week 31 and 35 observed
in 2014 did not occur in 2015.
This could be due to improvements in the spray programmes
leading to delayed capture. However
effectiveness of the spray programme or perhaps the reduction in
fruit availability increased the
desirability of the traps leading to an increase in D. suzukii
from August onwards. Numbers peaked
at the beginning of November, resulting in a higher trap catches
than in 2014 (Fig. 1.2.2).
-
24
Figure 1.2.3. Number of male and female D. suzukii trap catches
at farms 1 and 2 in 2015.
The number of D. suzukii in both Farm 1 and 2 followed a similar
activity pattern with the first signs
of activity found at the beginning of the year. This activity
increased in early spring from 14 April
- 8 June. More pronounced activity occurred towards the end of
the summer from 27 July and this
continued into the winter (Fig. 1.2.3). The peak of activity at
both farms occurred at the beginning
of November. The larger winter morphs of D. suzukii were still
found in May.
Figure 1.2.4. Ratio of adult male to female D. suzukii collected
from Farm 1 and Farm 2.
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
09-Mar 09-Apr 09-May 09-Jun 09-Jul 09-Aug 09-Sep 09-Oct
09-Nov
Nu
mb
er
of
D.
suzu
kii
Farm 1 Male Farm 1 Female
Farm 2 Male Farm 2 Female
-1
0
1
2
3
4
5
16-Feb 16-Mar 16-Apr 16-May 16-Jun 16-Jul 16-Aug 16-Sep 16-Oct
16-Nov
Rat
io o
f M
ale
to
Fe
mal
e
Farm 1
Farm 2Males
Females
-
25
The combined ratio of males to females in 2015 for both farms
was 1.14 male to female,
comparable to the ratio of 1.13 found last year. When analysed
individually the numbers remain
fairly similar, with ratios of 1.15 for Farm 1 and 1.09 for Farm
2 (Fig. 1.2.4). Therefore, males
were to some extent more prevalent overall in the current
trapping devices.
In 2014, there were generally higher trap catches of males in
the winter months. In addition, males
were captured first in both 2013 (13 August for males compared
to 17 September for females) and
2014 (in 2014 males Farm 1 on 18 February compared to females on
25 February). In 2015 this
trend continued with more males being caught in the winter and
early spring (Fig. 1.2.4). The
increased proportion of females to males in the first week of
April coincided with the appearance
of the first fecund females (Fig. 1.2.11).
Habitat assessments
D. suzukii were caught in all trap locations within Farm 1 and 2
in 2015 (Fig. 1.2.5 & 1.2.6). The
highest numbers of D. suzukii were found within the woodland
areas on both farms. Numbers
were highest from mid-September and peaked in November but
woodland was also shown to be
the preferred habitat in the spring, from late March until
mid-May.
On the crops that were sampled, much higher numbers of D.
suzukii were caught in the cherry
orchard. In the cherry orchard the ratio of male to female D.
suzukii showed a preference towards
the males. Numbers of D. suzukii increased from 24 August and
peaked 9 September (Fig. 1.2.5).
On the 5 October numbers of D. suzukii caught in the traps in
the cherry orchard were overtaken
by those in the woodland habitat and increased exponentially
from the beginning of November and
peaked on 9 November. The number of D. suzukii decreased in the
cherry orchards once all the
leaves had fallen.
-
26
Figure 1.2.5. Mean numbers of D. suzukii captured per habitat
type at Farm 1.
Figure 1.2.6. Mean numbers of D. suzukii captured per habitat
type at Farm 2.
0
200
400
600
800
1000
1200
1400
1600
1800
16-Feb 16-Mar 16-Apr 16-May 16-Jun 16-Jul 16-Aug 16-Sep 16-Oct
16-Nov
Ave
rage
nu
mb
er
of
D.
suzu
kii p
er
hab
itat
typ
eWoodland
Hedgerow
Cherry
Strawberry
Grape
0
200
400
600
800
1000
1200
1400
1600
1800
2000
16-Mar 16-Apr 16-May 16-Jun 16-Jul 16-Aug 16-Sep 16-Oct
16-Nov
Me
an n
um
be
r o
f SW
D p
er
hab
itat
Raspberry Hedgerow
Woodland
-
27
Other drosophila
The combined ratio of D. suzukii to other drosophila on both
farms was 0.46. However, at the
peak of SWD activity on Farm 2 the ratio climbed to 5.48 D.
suzukii to other drosophila, so that D.
suzukii were the dominant catch (Fig. 1.2.7).
Figure. 1.2.7. Ratio of D. suzukii to other drosophila
Reproductive stage of females
Measurements of body length and wing length were taken for all
individuals at both farms (Fig
1.2.8 and Fig 1.2.9). At both Farms 1 and 2 the average body
length remained high up until May.
However, throughout the summer months the average body length
decreased on both farms. This
corresponded with a reduced average wing length on both farms
from May onwards. This reduced
size was maintained up until the most recent sampling date (02
November 2015). In 2014, a
similar trend was observed in the average body length on Farm 1.
However, at Farm 2 the size
increased throughout 2014. All reproductive stages were seen
during the growing season on both
farms in 2015 (Fig 1.2.10 and 1.2.11). The period of greatest
reproduction (egg laying) was
between April and October. This suggests a months delay in egg
laying compared to the previous
year (March and September).
-1
0
1
2
3
4
5
6
16-Feb 16-Mar 16-Apr 16-May 16-Jun 16-Jul 16-Aug 16-Sep 16-Oct
16-Nov
Rat
io o
f SW
D:O
the
r D
roso
ph
ila
Farm 1
Farm 2
-
28
D. suzukii from Farm 2 appeared to have a longer period of
fecundity than Farm 1 (April through
to November), a trend also found in the previous years results.
At Farm 1 the polytunnels on
cropping areas were removed on 10th August 2015, whereas the
polytunnels at Farm 2 continue
to remain in place. Furthermore, at Farm 2 waste raspberry
fruits had not been removed. This
may explain the continued fecundity late into the season.
Fig. 1.2.8. Body and wing length measurements from Farm 1
Fig. 1.2.9. Body and wing length measurements from Farm 2
-
29
Fig. 1.2.10. Fecundity of D. suzukii at Farm 1.
Fig. 1.2.11. Fecundity of D. suzukii at Farm 2.
-
30
Summary
Adult D. suzukii were caught thoughout the year and were the
most abundant insect caught in
the traps from September onwards, outnumbering all the other
bycatch combined.
Higher numbers of D. suzukii were caught in cherry orchards than
in other crops surveyed.
The number of D. suzukii decreased in the cherry orchards once
all the leaves had fallen.
There is a high ratio of females in the traps from the beginning
of April, which also
corresponded to an increase in fertility.
The winter morphs are still found until May.
References
Gerdeman, B. S. 2010. Biology and management of spotted wing
drosophila, Drosophila suzukii
(Matsumura) in small fruits in the Pacific Northwest. IOBC
Working Group “Integrated Plant
Protection in Fruit Crops” Workshop on integrated soft fruit
production, 20-23 September.
Budapest, Hungary (speaker).
Gerdeman, B. S. and L. K. Tanigoshi. 2012. Comparative
reproductive condition between
laboratory cultures and field collected spotted wing drosophila
in the Pacific Northwest.
IOBC Working Group “Integrated Plant Protection in Fruit Crops”
Workshop on integrated
soft fruit production, 7-12 October. Kusadasi, Turkey.
(Speaker)
Pétavy, G. Moreteau, B. Gibert, P. and David, J. R. 2002
Phenotypic plasticity of body
pigmentation in Drosophila: influence of a developmental
thermoperiodic regime in two
sibling species. Physiological Entomology, Volume 27, Pages
124-135.
-
31
Task 1.3. Establishing a population model for Drosophila
suzukii
Introduction
In order to have a better understanding of Drosophila suzukii as
a new invasive pest to the UK,
knowledge of the population dynamics throughout the season of
this insect is important. Currently
it is unclear how D. suzukii populations behave throughout the
year and at which time of the season
the adults are utilising different crops, such as cherries,
strawberries and raspberries. A population
model would help aid the understanding and prediction of D.
suzukii population dynamics. This
model could be used to predict the first egg laying of female D.
suzukii and identify different
intervals with a high egg laying pressure. These intervals would
be useful to enable growers to
time the application of crop protection product spray
programmes. Furthermore, a population
model could be used for a better understanding of the life cycle
of D. suzukii and to develop a
more accurate prediction tool for a more sustainable integrated
pest management.
Materials and methods
The population model uses an algorithm to calculate population
size. These calculations were
done in R (R Core Team, 2015) combined with R Studio (RStudio
Inc, Version 3.2.1, 18-06-2015)
as an integrated development environment. A mathematical
package, ‘stringi’ (Gagolewski and
Tartanus, 2015), was also used to execute the algorithm.
Weather Data
Before an algorithm can be calculated it needs input values. One
of the key parameters which
influence the outcome of our algorithm, D. suzukii population
growth, is the weather. The algorithm
uses weather data as a basis for almost all calculations. The
weather data used in this study was
provided by Agrii (Agrii intelligence, MetQuest Weather
Stations). All imported weather data were
collected from different weather stations and were checked for
errors made by the electronics of
the weather stations. The files were also checked for consistent
measurements during 15 minute
intervals. In preparation for further calculations each file
with weather data was provided with
cumulative degree days (DD) based on the temperature. The
calculation with cumulative DD was
used to eliminate temperature variances during the different
seasons and across multiple years
and places. It is a tool used to compare different
climatological circumstances. For this data the
DD were calculated with 10 and 30°C as a lower and upper limit,
above and below which D. suzukii
reproduction is ceases (Equation 1). The range between 10 and
30°C corresponds with the normal
development of D. suzukii (Kinjo and Kunimi, 2014). Each degree
above the lower limit was
multiplied by the interval time between two measurements. If the
temperature rose above the
upper limit of 30°C or dropped below the lower limit of 10°C
there were no DD added. The
cumulative DD is the sum of the previous measuring points and
the DD calculated at a given time.
-
32
This method of calculating cumulative DD was more accurate than
using a single or double sine
method with the average daily temperatures, more frequently used
in population models.
Equation 1. Calculation of DD. With Tx= air temperature on a
given Timex, Tlower=10°C,
Tupper=30°C. Because of the 15 minute intervals there was a
conversion factor needed to
convert the DD in minutes to a DD in days by multiplying by
1440.
𝐷𝐷 (days) = [(𝑇𝑥 − 𝑇𝑙𝑜𝑤𝑒𝑟) ∗ (𝑇𝑖𝑚𝑒𝑥 − 𝑇𝑖𝑚𝑒𝑥−1)] ∗ 1440
𝑖𝑓 𝑇𝑙𝑜𝑤𝑒𝑟 ≤ 𝑇𝑥 ≤ 𝑇𝑢𝑝𝑝𝑒𝑟 𝑒𝑙𝑠𝑒 𝐷𝐷(𝑑𝑎𝑦𝑠) = 0
-
33
The weather data consisted of 15 minute intervals and included
air temperature, relative humidity,
rain and solar radiation. Currently the model only uses the air
temperature to model the population
dynamics. Under the assumption that temperature has the major
influence on the development of
this invasive insect this parameter will have the greatest
impact in the model. In a later phase
relative humidity and rain could be implemented as these are
known to affect activity of the adults
(Tochen et al. , 2015). Potentially rain and dry conditions may
decrease activity and prevent mating
and thereby effecting the next generation.
Estimation of first egg laying
For a population to grow, eggs need to be laid and larvae
develop through all of the development
stages. Therefore, the time of the first egg laying is a
critical time point for the algorithm. The
estimation of the first egg laying in spring was found by
comparing the calculated cumulative DD
with ovary development. The time points of general ovary
development were selected as points of
the first egg laying.
Table 1.3.1. Stage of ovarian development determination key
Number Stage of development
1 No distinguishable ovaries when opened
2 Ovaries are distinguishable when abdomen opened but no eggs
within
3 Ovaries distinguishable full of eggs without filaments when
opened
4 Mature eggs with filaments
5 Ovaries with few mature eggs, many wrinkled, may look slightly
yellow
-
34
Figure 1.3.1. The five different ovary stages as described in
Table 1 (photographs by Bethan
Shaw)
The ovary development was monitored by EMR, at the two sites
used in the habitat monitoring, by
dissecting five female D. suzukii from each site each week and
assessing their ovaries. The
habitats were monitored with modified drosotraps (Biobest) with
Cha-Landolt bait. Each ovary
received a number based on the stage of development (developed
by Beverley Gerdeman USA,
Table 1.3.1).
To establish the time of the first egg laying after winter,
stage three and four were used (Figure
1.3.1). These stages represented the most likely start of laying
eggs in the monitored week. The
ovary data were collected over a period of two years.
Population calculations
The population model requires different inputs, shown in Figure
1.. The imported source file,
“DD_year_name.csv” data file, contained the audited weather data
file combined with the
calculated cumulative DD for each 15 minute interval. This data
file was the base for all further
calculations. Next to the weather data additional parameters
were required. The second data file
was a fixed data frame which consisted of the development and
mortality rate of the adult flies for
a range of temperatures
(“Development-MortalityRateMin.csv”).
-
35
Figure 1.3.2. Calculation diagram to predict the D. suzukii
population. On top of the diagram the
different input data sets are shown to start the model. The
actual population calculations are split
up in two groups. The first set of calculations has to be done
every 15 minutes. The other group
of calculations has to be done each day. The different
algorithms to calculate mortality rate and
development rate are based on Asplen et al. (2015) as well as
the age based egg laying.
-
36
The development rate of the eggs and pupae are dependent on
temperature and their calculation
is based on a function from Asplen et al. (2015). The mortality
rate is also temperature based.
Due to the lack of information about the age specific
oviposition the assumption was made that
the temperature had no influence on this parameter. Therefore
all calculations were executed with
age specific egg laying at 21°C found in Asplen et al. (2015).
The maximum age was set at 90
days for the population model.
The first egg estimation (Table 1.3.2) was used as a starting
point for the actual algorithm to
calculate the D. suzukii population (green area in Figure 1.).
The calculation algorithm was divided
into two types of calculations. The first set of calculations
was done every 15 minutes and
calculated the development of each egg for a given temperature.
The mortality rate for each age
category was also calculated every 15 minutes. The second
calculation set used the age category
of the female adults to calculate the eggs laid during the day
and placed the emerged adults in at
age one of the “ageCategory” data frame.
Throughout the calculations there were a couple of assumptions
made in order to simplify the
model. All the eggs laid by D. suzukii females were assumed to
emerge, because currently there
is no information available about how many eggs survive the
three larval stages and pupation
before becoming an adult. There is also a lack of information
about the pre-oviposition period after
females emerge from the egg. In the model it was assumed that
there was no ripening period of
the ovaries for egg laying after female emergence. Therefore
when the females emerge we
assumed they could lay eggs immediately. A third assumption was
made, based on field data
collected over three years in the UK at EMR, about the
proportion of male / female adults. The
proportion was assumed to be 50/50. Therefore, half of all the
eggs laid will become females.
-
37
Results
Estimation of first egg laying compared to cumulative degree
day
Table 1.3.2 shows the calculated estimation of the possible
first and second egg laying in
cumulative degree days (DD). These results are based on the
ovarian development dissection
done by EMR. The estimation for the second spike of egg laying
has a high variance, making it
difficult to pinpoint the exact cumulative DD around which
female D. suzukii lay their first eggs.
Furthermore, the lower threshold of the 95% significance
interval was also too early compared to
the assessments done by the dissections of the ovaries. The
earliest record of developed ovaries
was found at 25 cumulative DD.
Table 1.3.2. Estimation of first and second egg laying in
cumulative degree days (DD). The
model used the 95% significance interval of the first egg laying
to establish the time range of the
first egg laying.
DD (day) Mean S.E.
68% significance 95% significance
lower
threshold
upper
threshold
lower
threshold
upper
threshold
1st egg
laying 36,88 13,12 23,76 50,01 10,64 63,13
2nd egg
laying 88,82 42,42 46,40 131,25 3,975 173,67
The 95% significance interval was further used to calculate the
start of the female population of
D. suzukii; the three intersections between the mean, upper- and
lower threshold of the estimated
first egg laying and the cumulative DD for each farm used in the
national monitoring (Figure 1.3.2)
were used to establish the starting points.
-
38
Figure 1.3.2. First D. suzukii egg laying estimation compared to
cumulative degree days. The two
black horizontal lines are the upper and lower thresholds of the
95% significance interval (Table
1.3.2), the red horizontal line is the mean (Table 1.3.2). Graph
A. shows the calculated cumulative
DD (minutes) of 2014 for seven farms. Graph B. shows the
calculated cumulative DD (minutes) of
2015 for nine farms. Those farms are part of the national
monitoring of D. suzukii.
-
39
Population estimation
With the input data and calculation explained in the previous
paragraphs an output could be
generated. More precisely, the population growth over the year
could be modelled. Figure 1.3.3
shows the female population of D. suzukii throughout the year.
The graph compares the three
different start points as calculated in Table 1.3.2. The
earliest start point at 10.64 cumulative DD
shows the highest number of female D. suzukii at the end of the
year.
Figure 1.3.3. Female D. suzukii population graph during 2014 at
EMR. The different lines show
the female population at different starting points for the
algorithm. The start points were based on
the 95% significance interval of the estimation of the first egg
laying (Table 1.3.2)
The calculation algorithm used to calculate the population had a
very large output of female flies.
As shown in Figure 1.3.3 the population reached the number of 20
billion female D. suzukii.
Although not visible in Figure 1.3.3 this exponential growth
starts at the beginning of June which
is visible when looking at a smaller time range (Figure 1.3.4).
The calculations for 2015 will follow
later when the complete weather data set is available.
-
40
Figure 1.3.4. Female D. suzukii population at EMR in 2014
focused in on April to June. The
population is calculated with 36.88 cumulative DD as a starting
point.
Discussion
The population model constructed in this study serves as a good
starting point for the population
estimation of D. suzukii. It succeeds in pinpointing the time of
egg laying and the start of the
population growth. By making various assumptions, based on the
literature, the model is a simple
representation of the reality. The cumulative DD are a good
parameter to start the calculations and
estimate the first egg laying.
The moment of rapid growth can be pinpointed, giving important
information to the farmers. Based
on weather data the start of the exponential growth of a given
year can be extracted. By providing
this information to the farmers they could use control
measurements at the right time to ‘knock
back’ D. suzukii populations before they grow exponentially.
Furthermore the model is an important
step towards a clear life cycle follow-up of D. suzukii,
providing a useful tool for research.
Despite several good points, there is still room for some
improvements. The current population
model gives a good representation of the population dynamics but
is less accurate in the
estimation of the actual population. The population model gives
really high numbers of the female
D. suzukii population. Those high numbers are probably a result
of the assumed low mortality rate
and 100% survival rate from laid eggs to emerged adults. The
high numbers of D. suzukii could
also be explained by the assumption that the females lay eggs
immediately after they emerge from
their pupae. The important thing to conclude so far is the start
of the exponential growth of the
-
41
population around the beginning of June for 2014. By
ameliorating the assumptions of mortality
and survival rate the model will also be applicable for absolute
numbers. Other factors such as
parasitism and predation will be more difficult to incorporate
in the population model because there
is currently no data available.
The calculation algorithm is currently based on temperature
input with corresponding mortality and
development rates. These rates are based on the research of
Asplen et al. (2015). The different
rates discussed in this article are concise. To improve the
basis of this population model it is
important to do further research on the influence of
temperature, relative humidity and light
radiation on the life cycle of D. suzukii.
The estimation of the first egg laying consists of four data
points (two farms over two years).
Although these values give a good indication, the sample size of
ten flies for each week is relatively
small to draw firm conclusions. This means that the standard
error and significance intervals are
relatively large and can be reduced by extending the research
about the ovary development.
A comparison with true data is necessary to validate the model,
this could be done with national
and habitat monitoring data. However the traps currently used
are not optimal to accurately
measure the population. Later in the season, during the ripening
of different crops, the trap catches
are less accurate because of competition between the traps and
the ripening fruit. In this situation,
D. suzukii is probably more attracted to the ripening fruit than
to the traps. Until April it is possible
to use currently available trap catches as a key to estimate the
actual population with the help of
a conversion factor. The current traps measure mainly the
activity of adult D. suzukii flies and less
so the actual population. Therefore further research to improve
the attractants and traps needs to
be continued. They are especially necessary to get a more
accurate reading during the ripening
season of the different crops and to establish a more accurate
prediction of the infection pressure
during ripening of the crops. This is the time when an accurate
prediction is most needed for the
growers. The ultimate goal of the model is to predict the fruit
damage pressure with the help of
available weather data at a given time during the year.
By using the same traps across the UK all flies behave the same
towards the traps. Therefore the
trap catches can be compared and can be used to validate the
population model. In order to
continue the construction of this model it is best to continue
with the current traps to have
consistency across multiple years. Better traps can be
supplemented next to the current traps as
to generate data which can be also used in this model in the
future.
-
42
Further research
The population model in this form can give useful information.
There are some possible
improvements which can make it even more relevant. In the
current model there is only an
implementation of the temperature. Further on, other weather
parameters have to be implemented.
To further improve the population model lab test, field tests
are needed to establish more detailed
information about the mortality and development rate of D.
suzukii in different weather conditions.
In a next step, the model has to be verified by the trap catches
of the national monitoring and
habitat monitoring. The habitat monitoring is especially useful
because this is more detailed than
the national monitoring and also provides more information about
the displacement during the year
from the woods to the crops and back.
There is not much knowledge about the survival rate during the
winter period and the trigger of the
female ovary development in spring. It would be useful to know
if female D. suzukii winter forms
have spermathecae (receptacula seminis) to store male sperm
before or after the winter. At the
moment the assumption was made that females do go in winter form
without ovaries to save some
energy to survive the winter. But what mechanism do male D.
suzukii winter forms have to survive
winter conditions?
To conclude it can be said that the generated population model
is a good starting point for
estimating population dynamics. This gives valuable information
for growers to start their pest
management of D. suzukii at the right time. Together with the
above mentioned improvements this
model can become even more relevant and result in huge
advantages for the agricultural sector.
Summary
A population model was constructed using fertility studies,
climatic data and biological
parameters
This model gave good estimates for the start of egg laying and
population growth early in the
year.
Work is continuing to optimise this model for use later in the
growing season.
References
Asplen, M., Anfora, G., Biondi, A., Choi, D., Chu, D., Daane,
K., Gibert, P., Gutierrez, A., Hoelmer,
K., Hutchison, W., Isaacs, R., Jiang, Z., Kárpáti, Z., Kimura,
M., Pascual, M., Philips, C.,
Plantamp, C., Ponti, L., Vétek, G., Vogt, H., Walton, V., Yu,
Y., Zappalà, L., Desneux, N.,
2015. Invasion biology of spotted wing Drosophila (Drosophila
suzukii): a global
perspective and future priorities. Journal of Pest Science.
88(3), pp 469-494.
-
43
Gagolewski M. and Tartanus B., 2015. R package stringi:
character string processing facilities.
URL: http://stringi.rexamine.com/.
Kinjo, H. and Kunimi, Y., 2014. Effects of temperature of the
reproduction and development of
Drosophila suzukii (Diptera: Drosophilidae). Journal of Applied
Entomology and Zoology,
49, pp 297-304
R Core Team, 2015. R: A language and environment for statistical
computing. R Foundation for
Statistical Computing, Vienna, Austria. URL:
http://www.R-project.org/.
Tochen, S., Woltz, J., Dalton, D., Lee, J., Wiman, N. and
Walton, V., 2015. Humidity affects
populations of Drosophila suzukii (Diptera: Drosophilidae) in
blueberry. Journal of applied
entomology.
-
44
Task 1.4. Monitoring SWD larval infestations in early, mid and
late season cherry
varieties
Objective
To characterize seven sweet cherry cultivars (‘Merchant’,
‘Simone’, ‘Kordia’, ‘Kordic’, ‘Penny’,
‘Regina’ and ‘Sweetheart’) by phenological and fruit quality
aspects to determine the ‘window’ of
exposure to D. suzukii egg laying in the fruit. We aimed to
identify the time taken from white fruit
to picking.
Experiment 1
Seven commercial varieties were studied: ‘Merchant’, ‘Simone’,
(early cropping); ‘Kordia’, ‘Kordic’,
‘Penny’, (mid-late cropping); ‘Regina’ and ‘Sweetheart’, (late
cropping). Fruits were sampled from
a commercial plot (Univeg, 4 years old) and the strategic,
unsprayed, plot of cv. Penny and
Sweetheart in ‘Rookery’ field (6-7 years old) at EMR by kind
permission of Graham Caspell (EMR
Farm Manager) (Table 1.4.1).
Harvest in a commercial orchard can start anywhere from late
June to mid-July depending on
preceding temperatures. We began phenological records in the
third week of April and the fruit
quality assessment was in the second week of June. There were
five trees of each variety. All of
the trees were in the same block in the Univeg or strategic
orchard with the same planting distance
and pruning system. The stage of flower development was
determined once a week from 23 April
until fruit set. After this the fruit development stage was
checked twice a week. The following
quality traits were measured;
Size (diameter measured using Vernier callipers in mm)
Colour using colourimeter (Chroma meter CR-400/410,
Konica-Minolta) and the Ctfil colour
chart (allowing us to avoid a possible error to quantify the
fruit colour and the difference among
varieties).
Firmness. The FirmTech1 (BioWorks, Stillwater, Okla).
Brix (where possible) (Palette 7 digital refractometer).
D. suzukii emergence. Five intact fruit from each variety.
We also used other EMR data on emergence and larval extraction.
The extraction method used
for the larval extraction was the standard crush test with sugar
solution.
-
45
Preliminary studies monitoring emergence of D. suzukii from the
fruits showed that in almost all
cases the adults emerged within 2 weeks of picking. Using a
model of D. suzukii development
time published by Tochen & Walton (Oregon University) we
predicted approximately when the
eggs were laid in the cherries. With a regression analysis based
on this model we estimated the
number of days from egg to adult for each temperature. We used
the temperature in the field for
one week and the lab temperature for the week before the
emergence for each variety. Once we
had the average temperature for those two weeks, we were able to
predict the days from egg to
adult and relate this to the fruit quality traits on the day of
egg laying. We used the temperature
data collected from the data loggers in the protected cherry
crop and the EMR meteorological data
for the two varieties which were not grown under plastic.
-
46
Table 1.4.1. Stage and physical attributes of cherry varieties
as they develop
Variety Date Stage
Firmness
(g/mm)
Size
(mm)
Colour
(key) Colour Brix
Kordia 24-Apr Stamens/Open flower
06-May Open flower
29-Jun Open flower
13-May Petal fall
20-May Calyx fall
26-Jun Green fruit 498 25.0 1.0 1.2 9.1
03-Jul Egg laying 405 26.3 1.4 1.6 11.0
21-Jul Emergence 305 28.5 5.4 5.8 11.2
24-Jul Harvest 267 28.3 6.4 6.6 10.3
Korvic 24-Apr Open flower
06-May Petal fall
13-May Fruit set
20-May Calyx fall
16-Jun Green fruit 400 24.7 1.3 Green/1 6.4
30-Jun Egg laying 326 27.1 2.4 2.2 13
17-Jul Emergence 291 25.6 5.4 5.8 9.3
21-Jul Harvest 239 26.3 6.2 6.6 14.3
Merchant 24-Apr Open flower
06-May Petal fall
13-May Fruit set
20-May Calyx fall
23-Jun Emergence (other farms) 263 23.0 2.4 2.4 10.5
30-Jun Egg laying (EMR) 202 29.3 4.4 5.4 11.0
14- Jul Harvest 174 25.7 6.0 6.8 16.0
17-Jul Emergence (EMR) 177 24.3 6.0 6.2 16.8
Penny 24-Apr Flower buds /Stamens
29-Apr Stamens
06-May Open flower
13-May Fruit set
20-May Calyx fall
16-Jun Green fruit 979 25.3 Green Green
30-Jun Egg laying 402 25.3 1.2 Green/1 12.0
07-Jul Larvae (other farms) 314 27.5 3.4 4.2 19.5
10-Jul Larvae (EMR) 327 24.3 3.8 4.8 19.6
17-Jul Harvest (emergence EMR) 326 26.7 5.6 6.0 16.4
-
47
Table 1.4.1 cont... Stage and physical attributes of cherry
varieties as they develop
Variety Date Stage
Firmness
(g/mm)
Size
(mm)
Colour
(key) Colour Brix
Simone 24-Apr Open flower
06-May Open flower/Petalfall
13-May Fruit set
20-May Calyx fall
09-Jun Green fruit 601 21.6 White/1 Green
03-Jul Harvest 188 24.5 6.0 6.0 12.1
Sweethe
art 24-Apr Open flower
29-Apr Petalfall
13-May Calyx fall
23-Jun Green fruit 681 22.5 1.0 Green
03-Jul Egg laying (EMR) 465 27.4 1.0 1.0 14.3
21-Jul Emergence (EMR) 360 26.4 3.6 4.8 18.5
24-Jul
Optimal timing of
harvest 298 28.5 4.6 4.6 14.8
Regina 24-Apr Stamens
29-Apr Stamens/Open flower
06-May Open flower
13-May Petalfall
20-May Calyx fall
30-Jun Green fruit 460 22.8 Green/1 Green/1 7.2
24-Jul Harvest 218 29.0 5.6 6.8 15.1
Results
No D. suzukii were found in Simone (early variety) and Regina
(late variety) during this trial. D.
suzukii emerged from the other 5 varieties. The average firmness
and Brix from all varieties where
D. suzukii was able to lay eggs was 356 (g/mm) and 12.1 degrees
respectively (see Table 1.4.1
and 1.4.2 for individual varieties). There was a positive
relationship between Brix and the softening
of the cherry fruits (Fig. 1.4.1). D. suzukii was not found in
the early variety Simone (picked first
week of July), but was found in Merchant. However it was only
found in Merchant fruits after the
optimal time for harvest had passed and hence the fruits were
very soft. For Regina it was possible
that this variety was well protected because it was in a
commercial orchard receiving plant
protection products.
-
48
Table 1.4.2. Firmness and Brix when D. suzukii was first able to
lay eggs
in cherry
Variety Firmness Brix Colour Sprayed
Kordia mid-
late 405 11 1.4 Yes
Korvic mid-
late 326 12.7 2.4 Yes
Merchant early 202 11 4.4 Yes
Penny mid-
late 402 12
Green-
1.2 No
Sweetheart late 465 14.3 1.0 No
Figure 1.4.1. Relationship between firmness and Brix in the
cherry varieties and when D. suzukii
was first able to lay eggs.
By combining all of the data collected and comparing firmness
and colour (chart) for each variety
and each assessment day we could predict the cherry colour at
the mean firmness for first egg
laying. For a firmness value of 356 grams/mm the colour was
almost 3 in the Ctfil colour chart.
This is a red, however, the data is somewhat misleading as the
crops which were treated with
plant protection products had eggs laid in them much later than
varieties which were not sprayed
(Table 1.4.1). For example, Sweetheart and Penny had eggs in
them at the green to chart colour
1 stage. Hence applying the plant protection products delayed
the ability of D. suzukii to lay viable
eggs in cherry fruits.
y = 0.0081x + 9.2916R² = 0.3525
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
0 100 200 300 400 500
De
gre
es
Bri
x (s
uga
r co
nte
nt)
Firmness (grams/mm)
-
49
Figure 1.4.2. Relationship between colour and firmness of all
tested varieties in two different
orchards.
Figure 1.4.3. Regression analysis between Brix and firmness for
each variety.
If all varieties are plotted together for firmness and Brix
there is no clear relationship. However
some varieties e.g. Sweetheart, Korvic and Merchant showed a
stronger relationship between
these two variables than the other varieties tested. Eggs were
laid from a Brix level of 11 in most
varieties (Table 1.4.2). Brix levels are highly dependent on
cultivar, annual climatic conditions,
cultivation system and rootstock (Gongalves et al. ,2006: Usenik
et al. ,2010) so we can conclude
that protecting fruits from the white stage is a recommended
option for reducing fruit damage by
D. suzukii.
y = -0.0159x + 8.7824R² = 0.6039
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
0 100 200 300 400 500 600
Co
lou
r
Firmness (g/mm)
y = -0.0024x + 13.621R² = 0.0052
y = -0.0462x + 25.352R² = 0.6894
y = -0.0021x + 11.508R² = 0.0148
y = -0.0336x + 20.253R² = 0.6011
y = -0.0309x + 28.397R² = 0.3609
y = -0.0156x + 21.425R² = 0.6109
y = -0.0077x + 16.098R² = 0.0485
y = -0.0147x + 12.908R² = 0.1286
0.0
5.0
10.0
15.0
20.0
25.0
0 200 400 600 800
De
gre
es
Brx
(Su
gar
con
ten
t)
Firmness (g/mm)
Overall
Linear (Korvic)
Linear (Kordia)
Linear(Merchant)Linear (Penny)
-
50
The following illustrations give the time of D. suzukii egg
laying and appearance of each variety in
2015;
a)
b) c)
Figure 1.4.4. a) Phenology and D. suzukii activity on cv. Penny
in relation to Brix and firmness
and b) photographs of when D. suzukii was able to lay eggs (30
June) and c) 28 July, past harvest,
when there was obvious damage to the fruits.
0
5
10
15
20
25
0
200
400
600
800
1000
1200
16Jun
19Jun
23Jun
26Jun
30Jun
3 Jul 7 Jul 10Jul
14Jul
17Jul
21Jul
24Jul
28Jul
(Gra
ms/
mm
)
Brix
Firmness
Egg laying (EMR)
Larvae (other plots)
Harvest
Emergence at EMR (other farms)
-
51
a)
b)
Figure 1.4.5. a) Phenology and D. suzukii activity on cv.
Sweetheart in relation to Brix and firmness
and b) photographs of when D. suzukii was able to lay eggs (3
Jul).
0
2
4
6
8
10
12
14
16
18
20
0
100
200
300
400
500
600
700
800
23Jun
26Jun
30Jun
3 Jul 7 Jul 10 Jul 14 Jul 17 Jul 21 Jul 24 Jul 28 Jul
Brix
Firmness
Emergence at EMR
HarvestEgg laying
-
52
a)
b)
Figure 1.4.6. a) Phenology and D. suzukii activity on cv. Kordia
in relation to Brix and firmness
and b) photographs of when D. suzukii was able to lay eggs (3
Jul).
0
2
4
6
8
10
12
14
16
0
100
200
300
400
500
600
26 Jun 30 Jun 03 Jul 07 Jul 10 Jul 14 Jul 17 Jul 21 Jul 24 Jul
28 Jul
Brix
Firmness
Emergence at EMR
HarvestEgg laying
-
53
a)
b)
Figure 1.4.7. a) Phenology and D. suzukii activity on cv.
Merchant in relation to Brix and firmness
and b) photographs of when D. suzukii was able to lay eggs (30
Jun).
0
2
4
6
8
10
12
14
16
18
0
100
200
300
400
500
600
9 Jun 12Jun
16Jun
19Jun
23Jun
26Jun
30Jun
3 Jul 7 Jul 10Jul
14Jul
17Jul
Brix
Firmness
Egg laying in other farms (Green fruit in EMR)
Emergence in other farms
13th of July fruit was
picked
Emergence at EMR
Egg laying at EMR
-
54
Fig. 1.4.8. Phenology and D. suzukii activity on cv. Simone in
relation to Brix and firmness
Fig. 1.4.9. Phenology and D. suzukii activity on cv. Regina in
relation to Brix and firmness
0
2
4
6
8
10
12
14
0
100
200
300
400
500
600
700
9 Jun 12 Jun 16 Jun 19 Jun 23 Jun 26 Jun 30 Jun 3 Jul
Brix
Firmness
0
2
4
6
8
10
12
14
16
0
50
100
150
200
250
300
350
400
450
500
30 Jun 3 Jul 7 Jul 10 Jul 14 Jul 17 Jul 21 Jul 24 Jul 28 Jul
Brix
Firmness
Harvest
Harvest
-
55
Fig. 1.4.10.a) Phenology and D. suzukii activity on cv. Korvic
in relation to Brix and firmness and
b) photographs of when D. suzukii was able to lay eggs (28
Jun)
Conclusions
The average firmness and Brix from all varieties when D. suzukii
was first able to lay
eggs was 356 (g/mm) and 12.1 degrees respectively.
No larvae were found in the early varieties when harvested at
optimal time.
References Usenik, V., Fabcic, J., & Stampar, F. (2008).
Sugars, organic acids, phenolic composition and
antioxidant activity of sweet cherry (Prunus avium L.). Food
Chemistry, 107, 185–192.
0
2
4
6
8
10
12
14
16
0
50
100
150
200
250
300
350
400
450
16Jun
19Jun
23Jun
26Jun
30Jun
3 Jul 7 Jul 10Jul
14Jul
17Jul
21Jul
24Jul
Brix
Firmness
Fruit was harvested 23rd of July
Emergence at EMR
Egg laying
-
56
Goncalves, B., Silva, A. P., Moutinho-Pereira, J., Bacelar, E.,
Rosa, E., & Meyer, A. S. (2007).
Effect of ripeness and postharvest storage on the evolution of
colour and anthocyanins in
cherries (Prunus avium L.). Food Chemistry, 103, 976–984.
Tochen, S., D.T. Dalton, N. Wiman, C. Hamm, P.W. Shearer, V.M.
Walton. 2014. Temperature-
related development and population parameters for Drosophila
suzukii (Diptera:
Drosophilidae) on cherry and blueberry. Environmental Entomology
43 (2), 501-510.
-
57
Experiment 2
Objective
To monitor the occurrence of D. suzukii larvae in early, mid and
late season cherry crops at 4
commercial farms in Kent UK as the fruit ripens. Weekly samples
were taken from an early, mid
and late variety at each site from the early white to fully ripe
fruit stages.
Methods
Four farms were chosen in the South East of England. There are
no experimental treatments.
The samples of fruit were taken from the growers commercially
sprayed cherry orchards.
Adult activity
Two standard D. suzukii adult monitoring traps (Biobest modified
traps with Cha-Landolt lure
system) were deployed in each orchard, one in the middle one at
the edge near to possible sources
of infestation, at the end of blossom (late April/early May).
These were monitored weekly by
grower staff for numbers of male and female D. suzukii in each
trap each week. Trap contents
were strained from the drowning solution through a paint filter,
which was then transferred to a
polythene bag, sealed and labelled with pencil (label in bag)
and sent to East Malling for
confirmation of counts.
Larval infestation of fruits
Four replicate samples of 50 fruits were taken from each variety
weekly from early white fruit to
harvest and beyond if possible. Samples were held in ventilated
plastic boxes (supplied by EMR).
Two samples were taken from the edge of the orchard close to
woodland or sources of natural
infestation and two samples were taken from the centre of the
orchard. Samples were taken
randomly, selecting the ripest (most forward) at the time.
Crush testing
Standard sugar method (see Objective 3).
Emergence testing
Each sample was held in a ventilated Perspex plastic box
(228x121x66 mm) covered with fine
mesh (300 mmx400 mm) held tightly in place with electrical tape.
Fruit was placed on paper towel
to absorb liquid. Boxes were held at room temperature at the
host farm then delivered to reception
at EMR where they were held at 20°C for 3 weeks. They were
assessed weekly for adult
-
58
emergence. Communication of results was by email between the
nominated person at each site
and NIAB EMR.
Results
No D. suzukii emerged from fruits between 14 May and 8 June
(Table 1.4.3). Low numbers were
detected in early varieties on all but Farm 3 (Merchant and
Burlat) from 15 June (
-
59
Ta
ble
1.4
.3.
Ta
ble
of
infe
ste
d s
am
ple
s o
f 5
0 c
he
rrie
s f
rom
co
mm
erc
ial c
he
rry f
arm
s.
E=
ad
ult
D.
su
zu
kii e
merg
ing
fro
m f
ruit
wit
hin
3 w
eek
s.
L=
nu
mb
ers
of
larv
ae e
xtr
acte
d u
sin
g s
ug
ar
flo
tati
on
me
tho
d. G
rey a
rea
s i
nd
icate
pic
kin
g h
ad
cea
se
d.
-
60
Task 1.5. Identify the common wild host plants of D. suzukii
adults and larvae in the
UK. Years 2-3.
See year 2 report task 1.3 for materials and methods.
Results
No choice emergence
D. suzukii that were introduced into boxes with fruits had a
second generation emerge from several
species.
Table 1.5.1. Field collected fruits exposed to adult male and
female D.
suzukii and resulting second generation emergence (no
choice).
Description Total D. suzukii
(per g fruit)
Wall Cotoneaster (Cotoneaster horizontalis) 2.9
Fig (Ficus carica) 2.3
Mistletoe (Viscum album) 1.5
Honeysuckle (Lonicera) 0.7
Elderberry (Sambucus) 0.7
Japanese rose (Rosa rugosa) 0.7
Nightshade (Solanum) 0.6
Yew (Taxus) 0.5
Rowan (Sorbus sp.), 0.3
Pink Pagoda (Sorbus hupehensis) 0.3
Spindle (Euonymus europaeus) 0.2
Guelder Rose (Viburnum opulus) 0.1
Hawthorn (Crataegus sp.) 0
Ivy (Hedera helix) 0
Damson (Prunus domestica subsp. insititia) 0
Dogwood (Cornus sanguinea) 0
Holly (Ilex aquifolium) 0
Pyracantha (Pyracantha sp.) 0
Rubella (Skimmia japonica) 0
Mahonia 0
-
61
Natural emergence
In 2014 D. suzukii naturally emerged from cherry, strawberry,
blackcurrant, black bryony and yew.
It did not emerge from cotoneaster, snowberry, guelder rose,
dogwood, hawthorn, red bryony and
rose in the samples we collected (Table 1.5.2). Natural
emergence was not observed from the
strawberry tree plant berries (Arbutus unedo), natural emergence
tests on mistletoe (Viscum
album) were not possible because wild fruits have not been found
(birds). Emergence was
observed from mistletoe (Viscum album) fruit collected from the
wild (Briem 2015) and may be a
first spring host in Central Europe. Lee et al. (2015) provide
further information on wild hosts of
D. suzukii.
-
62
Table 1.5.2. First date of collection resulting in natural
emergence.
Fruit Date Number emerged
Cherry 30 May 10
Blackberry 24 Jul 1
Strawberry 04 Aug 2
Raspberry 07 Aug 2
Blackcurrant 22 Aug 1
Elderberry 05 Sep 1
Black Bryony 13 Oct 8
Yew 04 Nov 6
Conclusions
Black bryony and yew berries may provide egg laying sites
towards the end of the year.
Cotoneaster, snowberry, guelder rose, dogwood, hawthorn, red
bryony and rose do not appear to
support D. suzukii development.
References
Briem M., Breuer K., Köppler H., Vogt F., 2015 Phenology and
occurrence of Spotted Wing
Drosophila in Germany and case studies for its control in berry
crops. IOBC Bull. Working
Group “Integrated Protection of Fruit Crops, Subgroup Soft
Fruits”, Vol. 109: 233-237.
Lee, J. C., Dreves, A. J., Cave, A. M., Kawai, S., Isaacs, R.,
Miller, J. C. et al (2015). Infestation
of Wild and Ornamental Noncrop Fruits by Drosophila suzukii
(Diptera: Drosophilidae).
Annals of the Entomological Society of America, 108(2),
117-129.
doi:10.1093/aesa/sau014
-
63
Task 1.6. Explore SWD overwintering and determine whether SWD
overwinters in
UK fruit crops, including dead plant material and polytunnel
structures (EMR; [JHI];
Years 3-4)
Raspberry cuttings were collected at 4 different sites. They
were placed in polythene bags with a
precision monitoring trap for 3 weeks (Fig. 1.6.1). The traps
were checked on a weekly basis for
D. suzukii. None were found on any of the sites, however small
numbers of anthocorids were
observed to emerge. We are currently trying different methods
(bags and cages) of extracting
